Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Bug] Package tools on jetson generate X86_64 wheel which cannot be installed #2493

Closed
3 tasks done
roman-duris opened this issue Oct 16, 2023 · 2 comments · Fixed by #2497
Closed
3 tasks done

[Bug] Package tools on jetson generate X86_64 wheel which cannot be installed #2493

roman-duris opened this issue Oct 16, 2023 · 2 comments · Fixed by #2497

Comments

@roman-duris
Copy link
Contributor

Checklist

  • I have searched related issues but cannot get the expected help.
  • 2. I have read the FAQ documentation but cannot get the expected help.
  • 3. The bug has not been fixed in the latest version.

Describe the bug

When Using package tools on Jetson Xavier AGX, the python wheels for mmdeploy runtime build withou error but cannot be installed due to bad patform flag that specifies manylinux2014_x86_64.

I have fixed this by changing if statement starting on line 78 in generate_config.py from:

if args.system in ['linux', 'jetson']:
        config['PLATFORM_TAG'] = 'manylinux2014_x86_64'
    else:
        config['PLATFORM_TAG'] = get_platform().replace('-',
                                                        '_').replace('.', '_')

to:

if args.system in ['linux' ]:
        config['PLATFORM_TAG'] = 'manylinux2014_x86_64'
    elif args.system in [ 'jetson']:
        config['PLATFORM_TAG'] = 'any'
    else:
        config['PLATFORM_TAG'] = get_platform().replace('-',
                                                        '_').replace('.', '_')

Reproduction

build mmdeploy and mmdetection from source, build SDK as described in build for jetson tutorial.
then
python tools/package_tools/generate_build_config.py --backend 'trt' --system jetson --build-mmdeploy --device cuda --build-sdk --build-sdk-monolithic --build-sdk-python --output config.yml --pplcv-dir $PPLCV_DIR
and
python tools/package_tools/mmdeploy_builder.py --config config.yml --output-dir pack

Environment

10/16 09:05:37 - mmengine - INFO - 

10/16 09:05:37 - mmengine - INFO - **********Environmental information**********
10/16 09:05:40 - mmengine - INFO - sys.platform: linux
10/16 09:05:40 - mmengine - INFO - Python: 3.8.13 | packaged by conda-forge | (default, Mar 25 2022, 05:56:18) [GCC 10.3.0]
10/16 09:05:40 - mmengine - INFO - CUDA available: True
10/16 09:05:40 - mmengine - INFO - numpy_random_seed: 2147483648
10/16 09:05:40 - mmengine - INFO - GPU 0: Xavier
10/16 09:05:40 - mmengine - INFO - CUDA_HOME: /usr/local/cuda-11.4
10/16 09:05:40 - mmengine - INFO - NVCC: Cuda compilation tools, release 11.4, V11.4.239
10/16 09:05:40 - mmengine - INFO - GCC: gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
10/16 09:05:40 - mmengine - INFO - PyTorch: 1.11.0
10/16 09:05:40 - mmengine - INFO - PyTorch compiling details: PyTorch built with:
  - GCC 9.4
  - C++ Version: 201402
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - LAPACK is enabled (usually provided by MKL)
  - NNPACK is enabled
  - CPU capability usage: NO AVX
  - CUDA Runtime 11.4
  - NVCC architecture flags: -gencode;arch=compute_72,code=sm_72;-gencode;arch=compute_87,code=sm_87
  - CuDNN 8.3.2
  - Build settings: BLAS_INFO=open, BUILD_TYPE=Release, CUDA_VERSION=11.4, CUDNN_VERSION=8.3.2, CXX_COMPILER=/usr/bin/c++, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DLIBKINETO_NOCUPTI -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Werror=cast-function-type -Wno-stringop-overflow, FORCE_FALLBACK_CUDA_MPI=1, LAPACK_INFO=open, TORCH_VERSION=1.11.0, USE_CUDA=ON, USE_CUDNN=ON, USE_EIGEN_FOR_BLAS=ON, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=OFF, USE_MKLDNN=OFF, USE_MPI=ON, USE_NCCL=0, USE_NNPACK=ON, USE_OPENMP=ON, USE_ROCM=OFF, 

10/16 09:05:40 - mmengine - INFO - TorchVision: 0.11.0a0+fa347eb
10/16 09:05:40 - mmengine - INFO - OpenCV: 4.7.0
10/16 09:05:40 - mmengine - INFO - MMEngine: 0.7.1
10/16 09:05:40 - mmengine - INFO - MMCV: 2.0.0
10/16 09:05:40 - mmengine - INFO - MMCV Compiler: GCC 9.4
10/16 09:05:40 - mmengine - INFO - MMCV CUDA Compiler: 11.4
10/16 09:05:40 - mmengine - INFO - MMDeploy: 1.0.0rc3+0196cd0
10/16 09:05:40 - mmengine - INFO - 

10/16 09:05:40 - mmengine - INFO - **********Backend information**********
10/16 09:05:41 - mmengine - INFO - tensorrt:    8.4.0.11
10/16 09:05:41 - mmengine - INFO - tensorrt custom ops: Available
10/16 09:05:41 - mmengine - INFO - ONNXRuntime: None
10/16 09:05:41 - mmengine - INFO - pplnn:       None
10/16 09:05:41 - mmengine - INFO - ncnn:        None
10/16 09:05:41 - mmengine - INFO - snpe:        None
10/16 09:05:41 - mmengine - INFO - openvino:    None
10/16 09:05:41 - mmengine - INFO - torchscript: 2.0.0+nv23.5
10/16 09:05:41 - mmengine - INFO - torchscript custom ops:      NotAvailable
10/16 09:05:41 - mmengine - INFO - rknn-toolkit:        None
10/16 09:05:41 - mmengine - INFO - rknn-toolkit2:       None
10/16 09:05:41 - mmengine - INFO - ascend:      None
10/16 09:05:41 - mmengine - INFO - coreml:      None
10/16 09:05:41 - mmengine - INFO - tvm: None
10/16 09:05:41 - mmengine - INFO - vacc:        None
10/16 09:05:41 - mmengine - INFO - 

10/16 09:05:41 - mmengine - INFO - **********Codebase information**********
10/16 09:05:41 - mmengine - INFO - mmdet:       3.0.0
10/16 09:05:41 - mmengine - INFO - mmseg:       None
10/16 09:05:41 - mmengine - INFO - mmcls:       None
10/16 09:05:41 - mmengine - INFO - mmocr:       None
10/16 09:05:41 - mmengine - INFO - mmedit:      None
10/16 09:05:41 - mmengine - INFO - mmdet3d:     None
10/16 09:05:41 - mmengine - INFO - mmpose:      None
10/16 09:05:41 - mmengine - INFO - mmrotate:    None
10/16 09:05:41 - mmengine - INFO - mmaction:    None
10/16 09:05:41 - mmengine - INFO - mmrazor:     None

Error traceback

Command:
pip install pack/mmdeploy_runtime/mmdeploy_runtime_gpu-1.0.0rc3-cp38-none-manylinux2014_x86_64.whl 
Traceback:
ERROR: mmdeploy_runtime_gpu-1.0.0rc3-cp38-none-manylinux2014_x86_64.whl is not a supported wheel on this platform.
@irexyc
Copy link
Collaborator

irexyc commented Oct 17, 2023

@roman-duris

Hi, thanks for point this. Could you please pull a request to fix this error.

@github-actions
Copy link

This issue is marked as stale because it has been marked as invalid or awaiting response for 7 days without any further response. It will be closed in 5 days if the stale label is not removed or if there is no further response.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants